Implementing Schema Evolution in Data Warehouse through Complex Hierarchy Semantics
Kanika Talwar, Anjana Gosain · 2012
Data in a data warehouse is collected from several heterogeneous data sources under a unified format, which aims to provide strategic outcomes to the decision makers and facilitate pattern and trend analysis. These data sources are dynamic in nature, due t o ongoing transactions in an organization and ever changing requirements. This dynamic nature of the data warehouse has to be d ealt with evolution in the data warehouse schema in order to incorporate all the new changes and requirements. In data warehouse systems, the hierarchies play a very important role in processing and monitoring information. So in order to handle complex hierarc hies in case of data warehouse evolution, we have proposed evolution operators and certain constraints that need to be fulfilled for ensuring data integrity and schema correctness. This schema correctness in case of evolution is ensured through triggers. In this paper, we have considered a formal metamodel to model the constructs in data warehouse. Also the constraints and operators are defined using the Uni - level Description Language (ULD) and the Multilevel dictionary definition (MDD) approach. The ULD representation exhibits uniform formulation of data, schema and their interrelationships while the MDD structures provide a way for direct implementation in a relational database system.